Modular construction of extended DNA recognition surfaces: mutant DNA-binding domains of the 434 repressor as building blocks

Single-chain derivatives of the 434 repressor containing onewild-type and one mutant DNA-binding domain recognize the generaloperator ACAA–6 base pairs–NNNN, where the ACAAoperator subsite is contacted by the wild-type and the NNNNtetramer by the mutant domain. The DNA-binding specificitiesof several single-chain mutants were studied in detail and theoptimal subsites of the mutant domains were determined. Thecharacterized mutant domains were used as building units toobtain homo- and heterodimeric single-chain derivatives. TheDNA-binding properties of these domain-shuffled derivativeswere tested with a series of designed operators of NNNN–6base pairs–NNNN type. It was found that the binding specificitiesof the mutant domains were generally maintained in the new environmentsand the binding affinities for the optimal DNA ligands werehigh (with K_d values in the range of 10^–11–10^–10M). Considering that only certain sequence motifs in place ofthe six base pair spacer can support optimal contacts betweenthe mutant domains and their subsites, the single-chain 434repressor mutants are highly specific for a limited subset of14 base pair long DNA targets.     

A calmodulin-target peptide hybrid molecule with unique calcium-binding properties

This paper describes the production and properties of a hybridprotein comprising the full length of the Xenopus laevis cabnodulin(CaM) sequence, followed, through a gh/cylgh/dne linker, bythe 26-residue CaM-binding region of myosin light-chain kinase(M13). This hybrid molecule appears to have high thermal stability(T_m > 75°C in the presence of Ca²⁺) as well as unusualCa²⁺-binding properties: (i) a wide-range biphasic Ca²⁺-bindingresponse (extending over pCa 4.8-7.4) and (ii) a high apparentbinding constant (pCa_50% = 6.3, a 10-fold increase from thatof wild-type CaM). NMR and CD data indicate that the CaM-M13hybrid molecule exists in equilibrium in an approximate 1:1ratio between two major conformations, one of which is similarto the compact globular structure of the CaM-M13 complex [M.Daira,G.M.Clore, A.M.Gronenborn, G.Zhu, C.B.Klee and A.Bax (1992)Science, 256, 632-638] and the other to the dumbbell-like structureof the wild type CaM [Y.S.Babu, C.E.Bugg and W.J.Cook (1988)J. Mol. Biol., 204, 191-204]. The biphasic Ca²⁺-binding curvecan be interpreted using a linear combination of two Hill bindingcurves with significantly different dissociation constants (2x 10^-6 M and 8 x 10^-6 M), which can be attributed to the twoconformations in equilibrium. The present study has opened anavenue to engineer proteins with higher Ca²⁺-binding affinitiesusing the known CaM structures as a template Received April 23, 1993; revised August 5, 1993; accepted August 25, 1993.     

Molecular modeling of single polypeptide chain of calcium-binding protein p26olf from dimeric S100B()

P26olf from olfactory tissue of frog, which may be involvedin olfactory transduction or adaptation, is a Ca²⁺-binding proteinwith 217 amino acids. The p26olf molecule contains two homologousparts consisting of the N-terminal half with amino acids 1–109and the C-terminal half with amino acids 110–217. Eachhalf resembles S100 protein with about 100 amino acids and containstwo helix–loop–helix Ca²⁺-binding structural motifsknown as EF-hands: a normal EF-hand at the C-terminus and apseudo EF-hand at the N-terminus. Multiple alignment of thetwo S100-like domains of p26olf with 18 S100 proteins indicatedthat the C-terminal putative EF-hand of each domain containsa four-residue insertion when compared with the typical EF-handmotifs in the S100 protein, while the N-terminal EF-hand ishomologous to its pseudo EF-hand. We constructed a three-dimensionalmodel of the p26olf molecule based on results of the multiplealignment and NMR structures of dimeric S100B(ßß)in the Ca²⁺-free state. The predicted structure of the p26olfsingle polypeptide chain satisfactorily adopts a folding patternremarkably similar to dimeric S100B(ßß). Each domainof p26olf consists of a unicornate-type four-helix bundle andthey interact with each other in an antiparallel manner formingan X-type four-helix bundle between the two domains. The twoS100-like domains of p26olf are linked by a loop with no sterichindrance, suggesting that this loop might play an importantrole in the function of p26olf. The circular dichroism spectraldata support the predicted structure of p26olf and indicatethat Ca²⁺-dependent conformational changes occur. Since theC-terminal putative EF-hand of each domain fully keeps the helix–loop–helixmotif having a longer Ca²⁺-binding loop, regardless of the four-residueinsertion, we propose that it is a new, novel EF-hand, althoughit is unclear whether this EF-hand binds Ca²⁺. P26olf is a newmember of the S100 protein family.     

Control of antibody-antigen interaction using anion-induced conformational change in antigen peptide

The binding of a monoclonal antibody to an epitope peptide wascontrolled by the conformational change of the epitope peptideinduced by anions. We synthesized peptides in which the epitopesequence DTYRYI for the monoclonal antibody AU1 is located betweenamphiphilic peptides (KKLL)_n and (LLKK)_n. In the absence ofan appropriate anion, the peptide was in a random coil stateand the epitope was linear. In contrast, in the presence ofan appropriate anion, the peptide exhibited an anti-parallel-helical structure and the epitope was subsequently `bent'.In the presence of 41 µM triphosphate, the associationconstant between the antibody and the peptide was decreasedby one order of magnitude in the case of n = 3 and at leastthree orders of magnitude in the case of n = 4 or 5. Oligo-DNAs,as anions, dissociated the antibody–peptide complex, whereastriphosphate did not. The DNA concentrations required for 50%dissociation of the antibody–peptide complex at pH 7.5were 4x10^–8, 1x10^–7 and 6x10^–6 M for decamer,octamer and hexamer DNA, respectively.     

Expression of synthetic genes coding for completely new, nutritionally rich, artificial proteins

Synthetic genes (A, AB and AHB) constructed and cloned intopKK233-2 vector were recloned from the parent plasmid into thenew procaryotic expression vectors pGFY221N and pBIO52. GeneAF^–B (coding for all amino acids besides phenylalanine)was obtained by ‘cassette mutagenesis’ from geneAB. The plasmid pGFY221N was constructed from pGFY218L by replacingthe PstI by an NcoI site; plasmid pBIO52 was derived from pGFY221Nthrough replacing the 221-bp EoRl/NcoI fragment with a syntheticDNA segment of 52 bp representing the Escherichia coli atpEgene translational initiation region. The genes A, AB, AHB andAF^–B in the vector pGFY221N were expressed with a six-amino-acid-longleader sequence; in pBIO52 the genes were expressed directly.in vitro expression experiments were successful with all thegenes except with the AHB gene integrated into pGFY221N. Inthe E. coli minicell system expression was demonstrated withthe A gene in pGFY221N and the AF^–B and AHB genes in pBIO52.Complete translation of the expressed genes AB, AF^–B andAHB in either the in vitro or in vivo systems could be shownby using ³⁵S-labelled N-terminal methionine and C-terminal cysteine.Both amino acids occur only once in the peptide sequences.     

Dihydrofolate reductase: control of the mode of substrate binding by aspartate 26

The complex of Lactobacillus casei dihydrofolate reductase withthe substrate folate and the coenzyme NADP^* has been shown toexist in solution as a mixture of three slowly interconvertingconformations whose proportions are pH-dependent and which differin the orientation of the pteridine ring of the substrate inthe binding site. The Asp26 – Asn mutant of L. casei dihydrofolatereductase has been prepared by oligonucleotide-directed mutagenesisand studied by one-and two-dimensional ¹H-NMR spectroscopy.NMR studies of the mutant enzyme–folate–NADP^* complexshow that this exists to > 90% in a single conformation overthe pH^* range 5–7.1. The single conformation observedcorresponds to conformation I (the ‘methotrexate-like’conformation) of the wild-type enzyme–folate–NADP^*complex. These observations demonstrate that Asp26 is the ionizablegroup controlling the pH-dependence of the conformational equilibriumseen in the wild-type enzyme.     

Expression of mouse immunoglobulin light and heavy chain variable regions in Escherichia coli and reconstitution of antigen-binding activity

The expression of immunoglobulin heavy and light chain variableregions in the cytoplasm of Escherichia coli and formation ofa functional heterodimer has been demonstrated. Variable domainsequences were taken from the heavy and light chain cDNAs ofthe monoclonal antibody Gloop 2 and engineered for expressionin a dual origin expression vector. The engineered genes vhg2and vlg2 were separately subcloned into the vector, creatingtwo expression plasmids. Expression of the heavy and light chainvariable region genes (encoding 116 and 109 amino adds respectively)was investigated in eight E.coli strains; the polypeptides wererapidly degraded in a host strain optimized for expression andin E.coli strains deficient in the major protease La (lon^-).Accumulation was permitted in severely protease-deficient E.colihaving a defective heat-shock response. A lon^- mutation in thisgenetic background permitted even higher accumulation. Expressionlevels were 7 and 1% of total bacterial protein for light andheavy chain variable regions respectively. Expression of theheavy chain variable region gene was increased by includinga longer Shine–Dalgarno sequence. Similar constructionsin the light chain vector had no effect on expression levels.The insoluble variable region polypeptides were reconstitutedinto a heterodimer possessing the full antigen binding characteristicsof both the parent monoclonal antibody and its Fab fragment.     

Multimer formation as a consequence of separate homodimerization domains: the human c-Jun leucine zipper is a transplantable dimerization module

Human c-Jun and c-Fos leucine zipper domains were examined fortheir ability to serve as autonomous dimerization domains aspart of a heterologous protein construct. Schistosoma japonicumglutathione S-transferase (GST) was fused to recombinant Junleucine zipper (rJunLZ) and Fos leucine zipper (rFosLZ) domains.SDS–PAGE ‘snapshot’ analyses based on disulphidelinkage of monomers demonstrated the ability of rJunLZ to functionas a dimerization motif in a foreign protein environment. Sterichindrance prevented formation of rJunLZ–GST::rFosLZ–GSTheterodimers whereas rJunLZ–GST::rFosLZ and rJunLZ::rFosLZ–GSTformed readily. Furthermore, rJunLZ–GST generated homodimerssuggesting fusion protein heterodimers interact differentlyto homodimers. Gel filtration chromatography confirmed thatGST is a dimer in solution and that attachment of a leucinezipper domain allows further interactions to take place. Sedimentationequilibrium analyses showed that GST is a stable dimer (K_a >10⁶ M^-1) with no higher multimeric forms. rFosLZ–GST weaklyassociates beyond a dimer (K_a {small tilde}4x10⁵ M^-1) and rJunLZ–GSTassociates indefinitely (K_a {small tilde}4x10⁶ M^-1), consistentwith an isodesmic model of association. The interaction of theseleucine zippers independently of GST association demonstratestheir utility in the modification of proteins when multimerformation is desired.     

Point mutation of alanine (31) to valine prohibits the folding of reduced lysozyme by sulfhydryl--disulfide interchange

In the preceding paper in this issue, we described the overproduction of one mutant chicken lysozyme in Escherichia coil.Since this lysozyme contained two amino acid substitutions (Ala³¹ValandAsn¹⁰⁶Ser)in addition to an extra methionine residue at theNH₂-terminus the substituted amino acid residues were convertedback to the original ones by means of oligonucleotide-directedsite-specific mutagenesis and in vitro recombination. Thus fourkinds of chicken lysozyme [Met^–1 Val³¹Ser¹⁰⁶-, Met^–1Ser¹⁰⁶-,Met^–1 Val³¹-and Met^–1 (wild type)] wereexpressed in E. coli. From the results of folding experimentsof the reduced lysozymes by sulfhydryl-disulfide interchangeat pH 8.0 and 38°C, follow ed by the specific activity measurementsof the folded en zymes, the following conclusions can be drawn:(i) an extra methionine residue at the NH₂-terminus reducesthe folding rate but does not affect the lysozyme activity ofthe folded enzyme; (ii) the substitution of Asn¹⁰⁶ by Ser decreasesthe activity to 58% of that of intact native lysozyme withoutchanging the folding rate; and (iii) the substitution of Ala³¹Val prohibits the correct folding of lysozyme. Since the wildtype enzyme (Met^–1-lysozyme) was activated in vitro withoutloss of specific activity, the systems described in this study(mutagenesis, overproduction, purification and folding of inactivemutant lysozymes) may be useful in the study of folding pathways,expression of biological activity and stability of lysozyme.     

Intrahelical side chain-side chain contacts: the consequences of restricted rotameric states and implications for helix engineering and design

Intrahelical side chain–side chain (sc–sc) interactionsare assumed to play a crucial role in the formation and stabilityof -helices, yet it was found that only 37.2% of all helicalresidues are involved in such close contacts, assuming a specificminimum contact distance. The majority (58.0%) of these weredetected between residues with amino acid sequence spacing i,i + 4. The low frequency of intrahelical sc–sc contactswith sequence separations i, i + 1 and ii, i + 3, each observedwith only about one-third of the i, i + 4 counts, can be directlyand generally attributed to the absence of the g^- conformationin helices for the dihedral angle X₁- However, if it was assumedthat each side chain may maximally make only one sc–sccontact, as most commonly observed, the percentage of contactingpairs increased relative to the maximum possible pairs for agiven sequence spacing by a factor of {small tilde}4, e.g. from20.9 to 81.7% for i, i + 4 contacts. Stereochemical reasonsare also given for the observation that i, i + 3 contacts arecomposed largely of ion or polar pairs, while hydrophobic residuesdominate the i, i + 4 contacts. No significantly increased densityof intrahelical sc–sc contacts with increasing helix lengthwas found. Although there were generally fewer intrahelicalcontacts between buried helical residues when more contactswere made to the tertiary protein environment, the number ofintrahelical contacts did not increase with increasing solventexposure of the helices. Implications for helix design and thepacking of helices are discussed.     

Molecular dynamics study of Ca2+ binding loop variants of parvalbumin with modifications at the `gateway' position

The helix–loop–helix (i.e. EF-hand) Ca²⁺ ion bindingmotif is characteristic of a large family of high-affinity Ca²⁺ion binding proteins, including the parvalbumins and calmodulins.In this paper we describe a set of molecular dynamics computationson the major parvalbumin from the silver hake (SHPV-B). In allvariants examined, both whole protein and fragments thereof,the ninth loop residue in the Ca²⁺ binding coordination sitein the CD helix–loop–helix region (the so-called`gateway' residue) has been mutated. The three gateway mutationsexamined are arginine, which has never been found at the gatewayposition of any EF-hand protein, cysteine, which is the residueobserved least in natural EF-hand sites, and serine, which isthe most common (by far) non-acidic residue substitution atthis position in EF-hand proteins in general, but never in parvalbumins.Results of the molecular dynamics simulations indicate thatall three modifications are disruptive to the integrity of themutated Ca²⁺ binding site in the whole parvalbumin protein.In contrast, only the arginine and cysteine mutations are disruptiveto the integrity of the mutated Ca²⁺ binding site in the CDfragment of the parvalbumin protein. Surprisingly, the serinevariant of the CD helix–loop–helix fragment exhibitedremarkable stability during the entire molecular dynamics simulation,with retention of the Ca²⁺ binding site. These results indicatethat there are no inherent problems (for Ca²⁺ ion binding) associatedwith the sequence of the CD helix–loop–helix fragmentthat precludes the incorporation of serine at the gateway position.Since the CD site is totally disrupted in the whole proteinserine variant, this indicates that the Ca²⁺ ion binding deficienciesare most likely related to the unique interaction that existsbetween the paired EF-hands in the whole protein. Our theoreticalresults correlate well with previous studies on engineered EF-handproteins and with all of our experimental evidence on the silverhake parvalbumin.     

Effects of mutations in the calcium-binding sites of recoverin on its calcium affinity: evidence for successive filling of the calcium binding sites

A molecule of the photoreceptor Ca²⁺-binding protein recoverincontains four potential EF-hand Ca²⁺-binding sites, of whichonly two, the second and the third, are capable of binding calciumions. We have studied the effects of substitutions in the second,third and fourth EF-hand sites of recoverin on its Ca²⁺-bindingproperties and some other characteristics, using intrinsic fluorescence,circular dichroism spectroscopy and differential scanning microcalorimetry.The interaction of the two operating binding sites of wild-typerecoverin with calcium increases the protein's thermal stability,but makes the environment around the tryptophan residues moreflexible. The amino acid substitution in the EF-hand 3 (E121Q)totally abolishes the high calcium affinity of recoverin, whilethe mutation in the EF-hand 2 (E85Q) causes only a moderatedecrease in calcium binding. Based on this evidence, we suggestthat the binding of calcium ions to recoverin is a sequentialprocess with the EF-hand 3 being filled first. Estimation ofCa²⁺-binding constants according to the sequential binding schemegave the values 3.7 x 10⁶ and 3.1 x 10⁵ M^–1 for thirdand second EF-hands, respectively. The substitutions in theEF-hand 2 or 3 (or in both the sites simultaneously) do notdisturb significantly either tertiary or secondary structureof the apo-protein. Amino acid substitutions, which have beendesigned to restore the calcium affinity of the EF-hand 4 (G160D,K161E, K162N, D165G and K166Q), increase the calcium capacityand affinity of recoverin but also perturb the protein structureand decrease the thermostability of its apo-form.     

Interactions of (Ala*Ala*Lys*Pro)n and (Lys*Lys*Ser*Pro)n with DNA. Proposed coiled-coil structure of AlgR3 and AlgP from Pseudomonas aeruginosa

The proteins, AlgR3 and AlgP, are involved in the regulationof alginate synthesis in Pseudomonas. They contain multiplerepeats of Ala^*Ala^*Lys^*Pro as do several other proteins thatresemble histones. The interactions of synthesis oligopeptidescomposed of repeated Ala^*Ala^*Lys^*Pro or Lys^*Lys^*Ser^*Pro unitswith DNA were studied by fluorescence of the Fmoc (9-fluorenylmethyloxycarbonyl)group attached to the N-termini of the peptides. DNA quenchingof the Fmoc fluorescence of the peptides was used to estimatethe apparent association constants for the interaction of Fmoc(AAKP)_nOH(n = 2, 4, 8, 18, 32) and of Fmoc(KKSP)_nOH (n = 2, 4, 8, 16,20, 32) with DNA. The Fmoc(AAKP)_nOH peptides bind to DNA onlyat low ionic strength; the Fmoc(KKSP)_n OH peptides interactwith DNA at both low (0.05 M KCl) and high (0.2 M KCl) saltAt low ionic strength an increase in the number of the repeatunits causes an increase in the apparent association constantup to {small tilde}2 x 10⁶ M^–1 for both types of peptidesat N 24. The insertion of an AAKTA unit into the middle ofthe Fmoc(AAKP)₈OH peptide increases its affinity to DNA. Wepropose a model of (AAKP)_n and of its interaction with DNA.The repeat unit consists of a single turn of -helix followedby a bend necessitated by Pro. The resultant coiled-coil formsa right-handed superhelix with 10 AAKPs per repeat distanceof {small tilde}33 Å. With only slight modification ofthe canonical parameters of this model the AAKP super helixfits into the major groove of B-form DNA with one AAKP tetramerper base pair repeat of 3.4 Å. The -amine nitrogen ofLys can form a polar hydrogen bond with a phosphate oxygen atomof the DNA backbone. A better fit is obtained when the modelis modified to accommodate [(AAKP)₅AAKTA]_n as actually observedin AlgR3. We suggest that this coiled-coil represents a generalmotif for other protein–DNA interactions.     

QSAR studies applied to the prediction of antigen-antibody interaction kinetics as measured by BIACORE

The objective of this work was to investigate the potentialof the quantitative structure–activity relationships (QSAR)approach for predictive modulation of molecular interactionkinetics. A multivariate QSAR approach involving modificationsin peptide sequence and buffer composition was recently usedin an attempt to predict the kinetics of peptide–antibodyinteractions as measured by BIACORE. Quantitative buffer–kineticsrelationships (QBKR) and quantitative sequence–kineticsrelationships (QSKR) models were developed. Their predictivecapacity was investigated in this study by comparing predictedand observed kinetic dissociation parameters (k_d) for new antigenicpeptides, or in new buffers. The range of experimentally measuredk_d variations was small (300-fold), limiting the practical valueof the approach for this particular interaction. However, themodels were validated from a statistical point of view. In QSKR,the leave-one-out cross validation gave Q² = 0.71 for 24 peptides(all but one outlier), compared to 0.81 for 17 training peptides.A more precise model (Q² = 0.92) could be developed when removingsets of peptides sharing distinctive structural features, suggestingthat different peptides use slightly different binding modes.All models share the most important factor and are informativefor structure–kinetics relationships. In QBKR, the measuredeffect on k_d of individual additives in the buffers was consistentwith the effect predicted from multivariate buffers. Our resultsopen new perspectives for the predictive optimization of interactionkinetics, with important implications in pharmacology and biotechnology.     

Mutant forms of {beta}-galactosidase with an altered requirement for magnesium ions

By random approaches we have previously isolated many variantsof Escherichia coli ß-galactosidase within a shortcontiguous tract near the N-terminus (residues 8–12 ofwildtype enzyme), some of which have increased stability towardsheat and denaturants. The activity of these mutants was originallyanalysed and quantitated in situ in activity gels without theaddition of magnesium ions to the buffer system. We now showthat the improved stability is only observable under such conditionsof limiting magnesium ion concentrations or in the presenceof appropriate concentrations of a metal chelator. In the presenceof EDTA, purified preparations of one of these mutant enzymeswere much more resistant to denaturants than wild-type, butthis differential was completely nullified in the presence of1 mM Mg²⁺. However, the stability of this mutant enzyme in EDTAwas lower than that shown by it, or the wild-type enzyme, inthe presence of magnesium ions. In addition, certain alterationswithin another N-terminal tract (residues 27–31 of wild-type)resulted in enzymes with greater dependence on Mg²⁺ than naturalß-galactosidase. We conclude that a small number ofresidue changes in a large protein can profoundly modulate therequirement for metal ion stabilization, allowing partial abrogationof this need in certain cases. Thus, some enzymes which requiredivalent metal ions for structural purposes only may be engineeredtowards metal independence.     

A pore-forming protein with a metal-actuated switch

Staphylococcal -hemolysin, a pore-forming exotoxin, is a polypeptideof 293 amino acids that is secreted by Staphylococcus aureusas a water-soluble monomer. It assembles to form hexameric poresin lipid bilayers. Previous studies of pore formation have establishedthe involvement of a central glycine-rich loop. Here, we showthat when five consecutive histidine residues replace aminoacids 130–134 at the midpoint of the loop, they providea switch with which pore activity can be (i) turned off by micromolarconcentrations of divalent zinc ions and (ii) turned back onwith the chelating agent EDTA. Planar bilayer recordings showthat Zn²⁺ and EDTA can act on open channels from either sideof the bilayer and thus demonstrate that the central loop linespart of the conductive pathway. Our results suggest that genetically-engineeredpore-forming proteins might make useful components of metalion sensors     

Bacterial secretion of the Fab fragment of a mouse monoclonal IgM that reacts with IgG variable regions

In this report, we describe the expression system that enabledus to produce in Escherichia coli the Fab fragment of a mouseIgM that has previously been shown to inhibit the binding ofIgG to autoantigens by interacting with their variable regions.In our system, both light chain and heavy chain fragments wereput under the control of the malE promoter. The light chainwas fused to the MalE signal sequence, while the heavy chainvariable and first constant region were fused to the alkalinephosphatase signal sequence. In this system, after inductionof the promoter with maltose, the Fab fragment could be detectedin a periplasmic extract of the bacteria by Western blottingand also by ELISA. This Fab fragment was purified on a goatanti-mouse immunoglobulin immunoadsorbent and biotinylated.The Fab fragment produced by E.coli reacted with the trinitrophenyl(TNP) hapten and F(ab')₂ fragments of mouse IgG and these reactivitiescould be specifically inhibited by the corresponding solubleantigens. The dissociation constants of this Fab were 1.65 x10^–6 M for TNP and 5 x 10^–6 M for IgG F(ab')₂ fragments,indicating that the affinity of the Fab fragment compared withthat of the whole IgM molecule was similar for TNP but was lowerfor IgG F(ab')₂ fragments